Table 1 Mechanical properties of rubberized concrete (28 days)
Properites Concrete 5% 10% 15% 20%
without rubber
lLB KLB lLB KLB lLB KLB ILB KLB
Compressive Strength (MPa) 45.69 41.71 42.49 33.69 37.30 24.75 26.96 22.14
23.91 Split Tensile Strength (MPa) 4.191 3.087 3.741 2.928 3.141 2.622 2.676 2.346
2.238 Unit Weight (kg/m') 2258 2190 2190 2120 2120 2050 2050 1980
1980
504 Mehmet Emlroglu and Servet YildlZ
2. Experimental procedure
The materials used in this study were fine aggregate, coarse aggregate, shredded rubber obtained from waste tires and cement. Fine crumb rubber and coarse crumb rubber replacing fine aggregate and coarse aggregate, respectively, were used to prepare the concrete of this study. Hybrid designs of the concrete and specific gravities of mix materials are considered in the calculations of properties.
Two groups of rubberized concrete mixtures were prepared;
.
ILB; the concrete in which certain percentages of fine rubber are replaced with fine aggregate.
KLB; the concrete in which certain percentages of coarse rubber are replaced with coarse aggregateWater/Binder ratio (0.55) and cement content (418 kg/m3) were kept eonstant in all samples. Four designated rubber contents were selected 5%, 10%, 15% and 20% by volume of total aggregate. To unify the rubber content, the range of rubber content was selected as total aggregate volume of plain concrete. Total 54 cube test specimens, <P150x300 mm, were prepared for compressive strength, split tensile strength and unit weight tests.
Some of the test results such as compressive strength, split tensile strength and unit weight of plain and rubberized concretes are listed in Table I.
SEM analysis was carried out in this study to examine the fractiJre and bond characteristics of rubber reinforced concrete. The SEM samplcs were collected from the fracture areas after the compressive strength tests. All samples chosen for the SEM analysis were coated with silver/gold for electrical conduction.
3. Experimental results and discussion
The rubberized concrete has to confirm celtain requirements for mechanical properties. They are particularly compressive and split tensile strengths. Although these values are considerably decrease with the addition of waste tire pieces, their values are still in rcasonable rangc. After the concrete
samples were selected for SEM studies, images were taken from each sample. A total of four
samples taken from center and edge of the concrcte cylinders in axial direction ,were studied. Fig. I shows the SEM image of nonnal concrete. It is clear that morphology of C-S-H gel appears as type III (denser-almost sphere) in the conventional concrete.
In the rubberized concrete, it is obvious that no interface bonding between cement paste and rubber tire has been maintained. An example of poor adhesion between them is shown in Fig. 2. Without an interface bonding, stress transfer between fibers and cement paste is possible owing to a
An investigation on its microstructure of'the concrete containing waste vehicle tire 505
Fig. 1 A SEM picture of fractured surface of control specimen
Fig. 2 lTZ between rubber and cement
mechanical interlocking. Seperation or breaking of the rubber was not oftenly observed on the fracture surface and generally the rubber appeared on the fracture sud'ace was in the pulled out form. No transition layer, or even trace of patch of tire material adhering to the interface, was observed. This suggests that the interfacial bonding strength is weak. Fig. 3 shows an example of pulled out a piece of rubber tire. As the rubber tires were being mixed, the hard particles of mix impacted and abraded the rubber surface as well as chopping procedure, causing deformation and so intrusions and extrusions. Grooves and pits also on the sud'ace of the rubber tire fibers, due to its
506 Mehmct Emiroglu and Servet YildiZ
.
Fig. 3 SEM image of pulled out rubber tire
Fig. 4 An image of ITZ between rubber tire, aggregate and cement pastc
chopped form, the paste and rubber tightly matched. Therefore, strong mechanical interlocking has been established and no dramatic drop on the bending strength is recorded for a certain volume fraction of rubber tire.
Intetface structure of three componcnts, cement pastc and aggregatc and rubber tire, is shown in Fig. 4. A strong interface bonding between cement paste and aggrcgate is establishcd but weak intelface bonding between rubber tire and aggregate and cement paste is obvious from the picture.
An investixation on its microstructure oj'the concrete containing waste vehicle tire 507
Fig. 5 Scanning electron microscopy picture of rubberized concrete
tire and cement paste in the concrete. Fig. 5 shows a micrograph of microcracks generated from ITZ between rubber tires and cement paste. It was found that these cracks usually start from lTZ between rubber tires and cement paste because of poor bonding characteristic around rubber tires and cement paste.' There are a lot of microcracks near ITZ in rubberized concrete. These microcracks seem clearly in Fig. 5.
In rubberized concrete, crack formation is different from plain concrete because bond strength between rubber and cement paste is poor than that of between aggregate and cement paste. Therefore initial cracks were formed around ruhber tires and cement paste in ruhherized concrete.
4. Conclusions
An experimental procedure was conducted which enabled the preservation of the compressive stress-induced microcracks and bonding characteristic in lTZ of mbberized concrete under applied load. Compressive strength and split tensile strength of the rubberized concrete is lower than traditional concrete because bond strength hetween cement paste and rubber tire particles is poor. Besides, pore structures in rubberized concretes are much more than traditional concrete.
Based on this study, the following conclusions can be said.
I. The ITZ characteristic of rubberized concrete is poor than the traditional concrete. Additionally, strength of a tire rubber is lower than that of traditional aggregate Due to these facts, compressi ve strength and split tensile strength of rubberized concrete is less than plain concrete.
2. There is systematic reduction in strength data with the increasing of the rubber content in traditional concrete. These reductions are related with the poor honding characteristic between rubbers and cement paste around the ITZ of rubberized concrete.
3. Although adhesion between the rubber and cement paste was weak, roughening interface is formed and it constructed a mechanical interlock which resisted relative movement of fibers
508 Mehmet Emiruglu and Servet YildiZ
immediately after cracks ini{jated.
4. C-S-H morphologies of normal and rubberized concrcte are same. From the SEM images, the addition of waste tire rubber in norIllal concrete has not any harmful effect on the C-S-H formation in concrete.
Acknowledgments
This study was partially supported by the Firat University Scientific ancl Technological Investigation CentTe.
References
Ncmati, K.M. (1997), "Fracture analysis of concrete using scanning electron microscopy", Scanning ]9(6), 426-430.
Di,unond, S. and Mindess, S. (1980), "A preliminary SEM study of crack propagation in mortar", Cement and Concrete Res., 10(4), 509-519.
G1ezie, P.I.P., MUller, A. and Roman, H.R. (2003), "Mierostrnetural investigation of silica fume-cement-lime mortar", Cement Concrete Compos., 25,171-175.
Nemati, K. M. (2000), "Preserving Microstmeture of concrete under load using the Wood's Metal Technique", Inl. .l. Rock Mech.Mining Sci., 37, 133-142.
Topeu, LB. (1997), "Assessment of brittleness index of rubberized concretes", Cement Concrele Res., 27(2), 177-183.
Topcu, LB. (1997), "Analysis of rubberized concrete as composite material", Cement Concrete Res. 27(8), 1135-1139.
Topeu, LB. (1997), "Collosion behaviours of rubberized concrete", Cement Concrele Res., 27(12), 1893-1898. Khatib, Z.K. and Bayomy, F (1999) "Rubberized pOitland cement concrete", J. Mala Civ. Eng., August, 11(3),
